OpenGAN: Open-Set Recognition via Open Data Generation

TL;DR

OpenGAN introduces a novel approach combining real outlier data and adversarially generated fake data, utilizing features from existing classifiers to improve open-set recognition beyond previous methods.

Contribution

The paper proposes OpenGAN, a new method that enhances open-set recognition by integrating real and synthetic outlier data with feature-based discriminators, outperforming prior approaches.

Findings

OpenGAN achieves state-of-the-art open-set recognition accuracy.

Combining real outliers with synthetic data improves discrimination.

Feature-based discriminator design enhances training stability and performance.

Abstract

Real-world machine learning systems need to analyze test data that may differ from training data. In K-way classification, this is crisply formulated as open-set recognition, core to which is the ability to discriminate open-set data outside the K closed-set classes. Two conceptually elegant ideas for open-set discrimination are: 1) discriminatively learning an open-vs-closed binary discriminator by exploiting some outlier data as the open-set, and 2) unsupervised learning the closed-set data distribution with a GAN, using its discriminator as the open-set likelihood function. However, the former generalizes poorly to diverse open test data due to overfitting to the training outliers, which are unlikely to exhaustively span the open-world. The latter does not work well, presumably due to the instable training of GANs. Motivated by the above, we propose OpenGAN, which addresses the limitation of each approach by combining them with several technical insights. First, we show that a carefully selected GAN-discriminator on some real outlier data already achieves the state-of-the-art. Second, we augment the available set of real open training examples with adversarially synthesized "fake" data. Third and most importantly, we build the discriminator over the features computed by the closed-world K-way networks. This allows OpenGAN to be implemented via a lightweight discriminator head built on top of an existing K-way network. Extensive experiments show that OpenGAN significantly outperforms prior open-set methods.